相关论文: DyePack: Provably Flagging Test Set Contamination …

Large language models are pre-trained on uncurated text datasets consisting of trillions of tokens scraped from the Web. Prior work has shown that: (1) web-scraped pre-training datasets can be practically poisoned by malicious actors; and…

Large Language Models (LLMs), such as ChatGPT and GPT-4, are designed to provide useful and safe responses. However, adversarial prompts known as 'jailbreaks' can circumvent safeguards, leading LLMs to generate potentially harmful content.…

Third-party resources ($e.g.$, samples, backbones, and pre-trained models) are usually involved in the training of deep neural networks (DNNs), which brings backdoor attacks as a new training-phase threat. In general, backdoor attackers…

The implications of backdoor attacks on English-centric large language models (LLMs) have been widely examined - such attacks can be achieved by embedding malicious behaviors during training and activated under specific conditions that…

Automatic LLM benchmarks, such as AlpacaEval 2.0, Arena-Hard-Auto, and MT-Bench, have become popular for evaluating language models due to their cost-effectiveness and scalability compared to human evaluation. Achieving high win rates on…

Large language models (LLMs) remain vulnerable to jailbreak prompts that are fluent and semantically coherent, and therefore difficult to detect with standard heuristics. A particularly challenging failure mode occurs when an attacker tries…

As software becomes increasingly complex and prone to vulnerabilities, automated vulnerability detection is critically important, yet challenging. Given the significant successes of large language models (LLMs) in various tasks, there is…

Existing studies on bias mitigation methods for large language models (LLMs) use diverse baselines and metrics to evaluate debiasing performance, leading to inconsistent comparisons among them. Moreover, their evaluations are mostly based…

The development of large language models (LLMs) depends on trustworthy evaluation. However, most current evaluations rely on public benchmarks, which are prone to data contamination issues that significantly compromise fairness. Previous…

With the rapid advancement of large language models (LLMs), ensuring their safe use becomes increasingly critical. Fine-tuning is a widely used method for adapting models to downstream tasks, yet it is vulnerable to jailbreak attacks.…

In recent years, foundation models (FMs) have solidified their role as cornerstone advancements in the deep learning domain. By extracting intricate patterns from vast datasets, these models consistently achieve state-of-the-art results…

We investigate security concerns of the emergent instruction tuning paradigm, that models are trained on crowdsourced datasets with task instructions to achieve superior performance. Our studies demonstrate that an attacker can inject…

Backdoor attack introduces artificial vulnerabilities into the model by poisoning a subset of the training data via injecting triggers and modifying labels. Various trigger design strategies have been explored to attack text classifiers,…

The use of large language models (LLMs) is widespread across many domains, including Software Engineering, where they have been used to automate tasks such as program generation and test classification. As LLM-based methods continue to…

Backdoor attacks, representing an emerging threat to the integrity of deep neural networks, have garnered significant attention due to their ability to compromise deep learning systems clandestinely. While numerous backdoor attacks occur…

Poisoning of data sets is a potential security threat to large language models that can lead to backdoored models. A description of the internal mechanisms of backdoored language models and how they process trigger inputs, e.g., when…

Although Large Language Models (LLMs) have demonstrated significant capabilities in executing complex tasks in a zero-shot manner, they are susceptible to jailbreak attacks and can be manipulated to produce harmful outputs. Recently, a…

Dataset contamination, where evaluation datasets overlap with pre-training corpora, inflates performance metrics and undermines the reliability of model evaluations. Measuring dataset contamination thus becomes essential to ensure that…

Unlike code completion, debugging requires localizing faults and applying targeted edits. We observe that frontier LLMs often regenerate correct but over-edited solutions during debugging. To evaluate how far LLMs are from precise…

We present MultiBreak, a scalable and diverse multi-turn jailbreak benchmark to evaluate large language model (LLM) safety. Multi-turn jailbreaks mimic natural conversational settings, making them easier to bypass safety-aligned LLM than…